diff --git a/blog/2025-07-29_llm-d-v0.2-our-first-well-lit-paths.md b/blog/2025-07-29_llm-d-v0.2-our-first-well-lit-paths.md
new file mode 100644
index 0000000..47df5ab
--- /dev/null
+++ b/blog/2025-07-29_llm-d-v0.2-our-first-well-lit-paths.md
@@ -0,0 +1,20 @@
+---
+title: "llm-d 0.2: Our first well-lit paths (mind the tree roots!)"
+description: Announcing the llm-d 0.2 release with new features and improvements that light the way forward for large language model deployment
+slug: llm-d-v0.2-our-first-well-lit-paths
+
+authors:
+  - robshaw
+  - smarterclayton
+  - chcost
+
+tags: [release, announcement, llm-d]
+---
+
+# llm-d 0.2: Our first well-lit paths (mind the tree roots\!)
+
+Our 0.2 release delivers three well-lit paths to accelerate deploying large scale inference on Kubernetes \- better load balancing, lower latency with disaggregation, and native vLLM support for very large Mixture of Expert models like DeepSeekR1.
+
+We’ve also enhanced our deployment and benchmarking tooling, incorporating lessons from real-world infrastructure deployments and addressing key antipatterns. This release gives llm-d users, contributors, researchers, and operators, clearer guides for efficient use in tested, reproducible scenarios.
+
+[rest of post to follow]
\ No newline at end of file